Estimating price rigidity in vertically differentiated food product categories with private labels

Relevance of the Topic: The rapid emergence of retail store brands, private labels (PL), over the past decades has created new and stiff competition for many established manufacturers of national grocery brands (NB). Apart from sizeable market shares in many staple food categories, retailers have successfully introduced new differentiated PL product lines to enter many higher quality segments (Dubois and Jódar-Rosell, 2010). PLs now account for 17% and 24% of the retail markets in the United States and Canada, respectively. Previous economic literature has addressed various issues and dimensions related to the competitive impact of PL including their economic significance to retail chains (Chintagunta et al. 2002), growth and development of PL product markets (Hoch and Banerji 1993), competitive interactions between PLs and NBs Cotterill et al. 2000; Bontemps et al. 2008; Karray and Herran 2009; Volpe 2010) and the use of private labels in exerting retail market power (Meza and Sudhir 2010). The economic literature has also taken great interest in retailer’s strategic use of PLs to counter the prior dominance of NB manufacturers (Richards, Hamilton and Patterson, 2010). However, the agri-food industrial organization literature has paid limited attention to the degree of price stability (rigidity) in the grocery-retailing sector given the otherwise high degree of price instability of many agricultural commodity and markets for intermediary goods. Of particular interest in this context has been the importance of the retail sales (discounting) phenomenon (Hosken and Reiffen 2001). Existing empirical studies show that many food products in retailing are characterized by relatively long periods of unchanged prices, followed by recurring periods of lower prices and after that a return to the initial level (Hosken and Reiffen 2001; Moeser 2002). Hosken, Matsa and Reiffen (2001) confirm that that within a product category the same items may be put on sale repeatedly, while others are rarely or never used in retail sales events. A number of possible triggers for price rigidity have been discussed. Differences in strategic management decisions between retail formats (Owen and Trzepacz 2002), the use of psychological pricing points (Slade 1998; Blinder et al. 1998; levy et al. 2011), and the economic literature evolving around concept of menu cost -the cost of changing retail shelf prices ((Levy et al. 1997; Owen and Trzepacz 2002). Despite the above evidence, still, empirical studies on price rigidity present in grocery retailing are underrepresented. And available evidence is mostly comprised of aggregate and largely descriptive analyses of cross-categorical data. Research Methodology: The objective of this paper to estimate the degree of price rigidity within vertically differentiated grocery-retail product categories that feature competing private labels and national brands. 1) To estimate price rigidity in categories where vertically differentiated PLs and NBs (basic, standard, premium) compete for consumer demand. The literature on retail private labels provides evidence of the increasing quality and differentiation of PLs used a strategic in the fight for market shares with established NBs in higher quality and premium product segments (e.g. organic). 2) To investigate the potential cost advantage if PLs and the impact of product-level retail margins as a determinant of price rigidity. In context of the increasing share of PLs bargaining power and PL retail margins have been discussed as potential contributors to retailer strategic pricing and promotional behavior. Available UPS-level wholesale price (price paid by the retailer) data enables us to analyze these components in the strategic uses of private labels and their potential impact on price rigidity across products and categories. 3) To test whether differences in underlying retailer management or distribution area (e.g. regional store management division), store size and configuration (e.g. store-storage area ratio), and store location (e.g. rural, urban, suburban) have an impact on price rigidity across vertically differentiated PLs and NBs. The analysis is based on a set of weekly store-level scanner data for a major North American retail chain with stores in the United States and Canada. Our data structure closely resembles the data set used by Eichenbaum, Jaimovich, and Rebelo (AER 2011) who assess the importance of nominal rigidities for retail pricing. The data is made available through the SIEPR-Giannini Data Center and covers 2004-2007 product-level sales in 200 UPC product categories for approximately 2,200 and 250 stores in the U.S. and Canada, respectively. Two case studies, salad dressing and packaged slide bacon, were chosen with regards to the paper’s focus of estimating rigidity among and between vertically differentiated PLs and NBs. The methodological framework is twofold. A regression approach used to explain price rigidity extends extents previous work by Herrmann et al. (2009) and Weber (2009). A second probabilistic model is used to estimate the probability of a price adjustment building on Kano’s (2007) paper. Potential for generating discussion during the meeting Our detailed case-study analysis of retail price rigidity and focus on vertically differentiated food–product categories reveals a number of interesting findings that go beyond existing research. Overall, our regression results indicate that retail price changes in both categories are not fully explained by retail sales and price jumps. We find some evidence of strategic retail behavior that deviates from the common ‘Hi-Lo’ retail price setting, where sales prices are followed by pre-promotion price levels. The results further reveal that price rigidity decreases as the share of discounts relative to the number of price observations increases. With sales increasing by 1%, price rigidity decreases by 0.5 %. We confirm the same qualitative effects for wholesale price adjustments on price rigidity. If the share of price wholesale price changes increases by 1%, price rigidity decreases by 0.035 %. Our results largely confirm previous results and offer several new contributions on the effect of wholesale price adjustments on retail price rigidity.